I am building my first site with Trellis + Bedrock + Sage.
Locally, I’m using Trellis. But remotely I have an extensive web server and separate database server running NGinx, HHVM and MySQL - among other things.
I am reading through the deployment docs for Trellis and I’m not seeing any information about nuances to publishing from a local Trellis install to a non-trellis install. For example, do I have to be running MariaDB remotely? Or will it be compatible with MySQL just fine?
Are there other things I need to prepare for or consider?
Okay, I’m ready to deploy my project to remote and here’s my questions regarding whether or not I should deploy to my existing setup or create a DO instance with Trellis.
My remote server is on AWS EC2 running NGinx, HHVM, among some other things. It’s got a specialized permission structure so I can give developers individual accounts to SSH / SFTP around the /home directory without breaking WP write permissions, and a handful of other bells and whistles.
I have a separate database server running MySQL and Reddis that isn’t public facing and only allows access directly from the other web server.
This all allows me to scale as we add websites by creating additional web servers without needing to keep MySQL on the same server.
Now, I like that Trellis uses MariaDB, and NGinx, etc… but I find it hard to believe it could stack up to what I’ve got going.
But I don’t have an Ansible Playbook for my setup, or the time, and probably the skills, to develop one. So I like that Trellis automates that deployment. That’s huge. But from a security standpoint I know my system and I’ve worked hard to lock it down using best practices and beyond without compromising ease of use for my staff (devs).
@ben or others, based on this information should I favor my setup or Trellis? I understand it may just be a matter of opinion, but opinions are welcome.
